Accurate Modeling of Water Clusters with Density-Functional Theory Using Atom-Centered Potentials.

Area of Science:

• Computational Chemistry
• Theoretical Chemistry
• Quantum Chemistry

Background:

• Density-functional methods often exhibit deficiencies in accurately modeling the binding energies of water clusters.
• Accurate prediction of water cluster energetics is crucial for understanding various chemical and physical processes.

Purpose of the Study:

• To develop and evaluate water-specific atom-centered potentials (ACPs) for improving density-functional theory (DFT) calculations of water clusters.
• To assess the accuracy of ACP-corrected DFT methods for predicting absolute and relative binding energies of water clusters of varying sizes.

Main Methods:

• Developed water-specific ACPs using accurate ab initio reference data.
• Employed the BHandHLYP density functional in conjunction with aug-cc-pVTZ basis sets, with and without dispersion corrections.
• Validated ACP performance on water clusters ranging from (H2O)10 to (H2O)17, and applied the method to (H2O)25.

Main Results:

• ACPs improved binding energy predictions for water clusters up to (H2O)10 to within 0.44 kcal/mol, a significant improvement over uncorrected DFT methods which showed errors up to 6 kcal/mol.
• ACP-based approaches accurately predicted binding energies for larger clusters ((H2O)16,17) within 0.3-2.2 kcal/mol of high-level ab initio results.
• The BHandHLYP/aug-cc-pVTZ-ACP method proved effective in identifying minimum-energy structures for (H2O)25.

Conclusions:

• Water-specific ACPs substantially enhance the accuracy of DFT calculations for water cluster binding energies, improving predictions by over an order of magnitude.
• The BHandHLYP/aug-cc-pVTZ-ACP approach offers a computationally affordable and accurate alternative to wave function theory methods for studying water clusters.
• ACPs are a valuable tool for improving the modeling of non-covalent interactions in molecular systems.